Advisory Opinion

Legal Vice Presidency
The World Bank

ADVISORY OPINION

ON

CERTAIN ISSUES ARISING IN CONNECTION
WITH RECENT SANCTIONS CASES

No. 2010/1

Anne-Marie Leroy
Senior Vice President and General Counsel

November 15, 2010

TABLE OF CONTENTS

I. Issues of General Relevance

A. Sources of Law for World Bank Sanctions

1
B. Rules of Evidence and Standards of Proof

10
C. Mens Rea as an Element of Sanctionable Practices

12
D. Theories of Liability

17

II. Issues Relating to Fraudulent Practice

A. Recklessness and Fraudulent Practice

22
B. Application of the Knowing and Reckless Standard to pre-2006 Definitions

31
C. Omission as an Element of Fraudulent Practice in pre-2004 Definitions

31
D. Establishment of Prospective Fraud in Bids

31

III. Other Issues

A. The Sanctioning of Government Officials

32

NOTES 34

ADVISORY OPINION ON CERTAIN ISSUES ARISING
IN CONNECTION WITH RECENT SANCTIONS CASES

Anne-Marie Leroy, Senior Vice President and General Counsel
November 15, 2010

1. The Legal Vice Presidency (LEG) has been asked by the Integrity Vice
Presidency (INT) to provide advice on certain legal issues that have arisen in connection
with recent sanctions cases. The Office of Evaluation and Suspension (OES) has also
welcomed more clarity on the legal standards that they should be applying the sanctions
cases, and has referred some additional issues for LEGs consideration. These issues
include a number of issues of general relevance to sanctions cases, including (1) the
sources of law for the Banks sanctions regime and their application to sanctions cases,
(2) the proper approach to the concept of mens rea as an element of sanctionable
practices, and (3) the proper standards for assessing evidence and the burden of proof.
We have also been asked to provide advice on a number of issues surrounding the proper
interpretation of fraudulent practice, including the proper understanding of the concept of
recklessness, and whether sanctions cases may be brought against government officials.

2. This Advisory Opinion articulates general principles that, in our view, should
guide the consideration of each of these issues in the context of sanctions cases governed
by the IBRD/IDA Sanctions Procedures. It is intended as a general reference for INT,
OES and the Sanctions Board in the preparation and disposition of future sanctions
cases.
1
We express no view as to the legal principles and standards that should be applied
to sanctions cases brought under the IFC or MIGA procedures.

I. !""#$" &' ($)$*+, *$,$-+).$

A. Sources of Law for World Bank Sanctions

3. Perhaps the most fundamental issue that arises in sanctions cases is what law
applies to the case. The matter is often straightforward, but can become complex when
two perceived sources of law (e.g., the Procurement Guidelines and the Standard Bidding
Documents) conflict. Our analysis begins with identifying the sources of law for
sanctions cases and the hierarchy of norms among them. In our view, the relevant sources
of law are the following, to be applied in the following order of precedence:

4. The Articles of Agreement. Though often overlooked in the context of specific
cases, we should not lose sight of the underlying legal basis for the sanctions regime,
which is the fiduciary duty to protect the use of Bank financing in the Articles of
Advisory Opinion No. 2010/1
November 15, 2010

2

Agreement.
2
This provision not only provides the legal basis for the regime but it also
delimits its scope. Thus, the Bank has not asserted the authority to debar parties based on
the misuse of project co-financing, nor has it debarred for wrongdoing that may be illegal
or immoral (such as tax evasion) but does not impinge on the use of Bank loans, credits
or grants. The fiduciary duty rarely affects the disposition of sanctions cases, because
INT is well aware of the limitations that the Articles impose on the regime. Nevertheless,
on occasion the tenuous connection of a case to the misuse of loan proceeds raises
constitutional issues of this kind.

5. The Policy Framework and Legislative History of the Sanctions Regime. The
policy framework for the Banks sanctions regime, which can be found in the Thornburgh
Reports,
3
2004 Sanctions Reform Board Paper,
4
subsequent Audit Committee papers,
5
in
the minutes of Board and Audit Committee meetings and in the various travaux
prparatoires for these policy statements, which we often refer to as the legislative
history for the sanctions regime, plays a multifaceted role in sanctions cases.

6. Most commonly, we turn to the legislative history not as an independent source of
law, but to help us interpret the meaning either of the Sanctions Procedures or other
elements of the legal framework, particularly the various definitions of fraud and
corruption (see below). For example, LEG has reviewed the legislative history in
advising that the pre-2004 definition of corrupt practice did not encompass secondary
forms of liability such as aiding and abetting.
6

7. More rarely but still pertinent in some cases, it is useful to bear in mind the policy
purposes that the sanctions regime was established to serve when looking at particular
issues that may arise in connection with individual cases or potential cases. In this sense,
the policy framework, as evidenced by the legislative history, can be seen as not simply a
source of interpretation, but as constituting an independent source of law second in
precedence only to the Articles themselves. At its most basic, the question should be
asked: was this the kind of conduct that the sanctions regime was established to address?
The issue can arise even in cases where the letter of the definitions of fraud and
corruption, which are stated in extremely broad terms, might be interpreted to fit the
conduct in question.

8. The Legal Framework for the Bank Financing. While the above sources of law
take precedence over all others, as mentioned they rarely have a direct effect on the
disposition of individual sanctions cases. The main source of law, and the one that
governs every sanctions case, is the legal framework for the Bank finance in connection
with which the alleged Sanctionable Practice took place. This is because it is the legal
framework that (i) establishes the Banks jurisdiction to sanction firms and (ii) contains
the relevant definitions of Sanctionable Practice by which the allegations that are the
subject to sanctions proceedings should be judged. A sanctions case stands or falls on
whether the alleged misconduct falls within at least one of these definitions.

9. Normally, the issue is straightforward. The Procurement, Consultant and/or Anti-
Corruption Guidelines apply to the Bank project or program in question, establishing the
November 15, 2010

3

sanctions regimes jurisdiction. The Statute of the Sanctions Board provides that the
Board will have the competence to hear cases and impose sanctions as set out in the
Sanctions Procedures;
7
the Sanctions Procedures provide that the cases subject to them
include cases involving Sanctionable Practices in connection with Bank financed or
Bank executed projects and programs governed by the Banks Procurement Guidelines,
Consultant Guidelines or Anti-Corruption Guidelines.
8
These same Guidelines contain
the applicable definition of Sanctionable Practices that is the legal standard against which
the alleged misconduct is to be assessed. Typically, sanctions cases involve bidding for a
Bank financed contract, so the Procurement Guidelines that apply to the project also
apply to the case, and the bidding documents under which the bidding took place reflect
the provisions of the applicable version of the Procurement Guidelines.

10. However, issues arise when there are variations between these documents, in
particular when the definitions in the Procurement Guidelines are at variance with those
in the bidding documents. This typically happens when the Bank has updated the
definitions of Sanctionable Practices in the Procurement Guidelines, and the Bank and
the borrower have agreed to use updated bidding documents that reflect the new
definitions, but the Loan Agreement is not (formally) amended to reflect the change. This
issue is discussed at some length below.

11. INT has sometimes asked why new versions of the definitions of Sanctionable
Practices should not apply, ipso facto, to new sanctions cases or, at least, to alleged
misconduct that occurs after their adoption. The answer lies simply in the fact that the
sanctions regime, at least as it is currently constructed, is based on a quasi-contractual
model: as stated above, the Banks jurisdiction (as stated in the Sanctions Procedures) is
established by application of the Procurement, Consultant or Anti-Corruption Guidelines
to a particular project.
9
This is typically accomplished by agreement between the Bank
and the relevant borrower in the legal agreement for the relevant loan and project. Under
current model forms, the legal agreement specifies that a particular version of the various
Guidelinesand therefore a particular version of the definitionswill apply to the loan
and project for their duration, unless the legal agreement is amended or, if the agreement
so stipulated, the parties otherwise agree. These are therefore the definitions that the
Bank has agreed to apply in judging any alleged misconduct that may occur under the
project in question.
10

12. It should be pointed out that this is not the only possible way in which the
sanctions regime could theoretically operate. For instance, legal agreements could
incorporate by reference the various Guidelines as amended from time to time, in which
case new versions of the definitions would apply automatically to any conduct that occurs
after their adoption. More radically, the argument could be made that the Bank has
inherent authority, by virtue of its fiduciary duty under the Articles of Agreement, to
sanction firms that have engaged in fraud and corruption in connection with its financing.
In theory, this is a legally sound position and was invoked at least once in the Banks
history to debar a firm before the sanctions regime was formally established in 1996.
11

Once the formal system was established, however, the Bank committed itself to
November 15, 2010

4

exercising its inherent authority in a particular fashion, so it is now constrained by the
legal framework outlined in this Opinion.

13. Authoritative Interpretation. Given that the formal legal framework for the
Banks sanctions regime is rather thin and the main legal standards to be applied to
sanctions cases, the various definitions of Sanctionable Practices, are notoriously broad, it
has always been recognized that they would require interpretation over time. Of course,
general principles applicable to the interpretation of legal texts should apply. Therefore,
the first and most important source of interpretation is the plain meaning of the text itself.
No interpretation may do violence to that plain meaning. But where the meaning is open
to variable readings, then it is fair and useful to turn to exogenous sources.

14. The most important source of interpretation is the Banks own legislative history,
as discussed above. Sometimes (but surprisingly seldom) the legislative history can
illuminate the intended meaning of a definition, a provision of the Sanctions Procedures
or some other aspect of the legal framework for sanctions cases, or provide the basis for a
purposive interpretation of the definition. Given the paucity of guidance in the legislative
history, however, ongoing guidance on interpretation is required. The first formal
guidance for IBRD/IDA sanctions cases
12
is contained in the Commentaries on the Anti-
Corruption Guidelines, which include interpretations of the 2006 definitions.
13
The
interpretations of the various Sanctionable Practices in the Commentaries will bear on our
advice in this Advisory Opinion.

15. Another source of interpretation, and one that the Bank intends to rely on heavily,
is the jurisprudence of the Sanctions Board. It was for this reason, at least in part, that the
Bank has decided to require fully reasoned, published Sanctions Board decisions. But the
Sanctions Board may not be in a position to provide the requisite guidance in all cases.
First, in practical terms, it will take some considerable time for the Board to develop
sufficient jurisprudence on a range of legal issues, so issues of first impression are likely
to arise for the foreseeable future. Second, the Sanctions Board was not established as a
policy-making body, so certain issues, in particular as to the proper interpretation of the
Banks legal and policy framework, lie outside its purview.

16. It is for these reasons that LEG has proposed to provide authoritative advice on
the proper interpretation of the Banks legal and policy framework, including the
Sanctions Procedures and the various definitions of Sanctionable Practices. LEGs
advice, however, must of necessity be limited to legal principles. It remains within the
purview of the EO and Sanction Board to apply the law to the particular facts of cases.
Moreover, since no general legal advice can or should attempt to cover all fact patterns,
and many legal standards are fact-dependent, these decision-makers play a key role in
developing jurisprudence over time that will articulate the specific standards by which
general legal principles are applied.

17. It is important to bear in mind that interpretation, as mentioned at the outset, must
be carried out according to general principles of legal interpretation, first and foremost
that any interpretation must flow logically from the intended meaning of the text.
November 15, 2010

5

Interpretation, even of texts as broadly stated as the definitions, must not slip into a
disguised form of amendmentwhich would amount to a retroactive application of
norms.

18. General Principles of Law. While there is no formal document in the legal
framework for the Banks sanctions regime that explicitly recognizes general principles
as a source of law for the Banks sanctions regime, it is not uncommon in international
14

and administrative law to resort to general principles to resolve legal issues not clearly
addressed within the applicable legal framework. LEG views general principles of law as
a useful and legitimate source of law for the sanctions regime, within certain parameters.
The formal legal framework for the Banks sanctions regime is rather thin and the main
legal standards to be applied to sanctions cases, the various definitions of Sanctionable
Practices, are notoriously broad.

19. The decision whether or not it is appropriate to apply a particular legal theory
based on general principles of law to sanctions cases must be grounded both in legal and
legal policy considerations. The threshold question has to be whether the purported
general principle of law actually exists as a matter of legal fact.
15
Before making its
argument based on a theory imported from general principles, INT (or any other party
seeking to assert or apply such a theory) should first establish, through appropriate
citations to national and international norms, the existence of the general principle in
question. We understand the potential difficulties here, but to be considered a general
principle, in our view, at a minimum, it must be shown that the legal theory in question is
supported by leading jurisdictions in both civil and common law systems.
16

20. Once the theory is established as a general principle as a matter of legal fact, the
decision whether or not it should be recognized as part of the Banks legal framework is
one of legal policy. LEG has proposed to assume a more proactive role in these decisions
because it is uniquely placed to do so, for a number of reasons, first of all because of its
institutional role as custodian of the Banks legal framework. It is uniquely able to assess
whether a particular legal theory would make sense within the larger framework, bearing
in mind, for example, that the definitions of sanctionable practices are also part of the
Banks operational legal framework (General Conditions, Procurement, Consultant and
Anti-Corruption Guidelines) and so their interpretation impacts on both Bank operations
and member country relations.

21. Moreover, the policy grounds for the application of general principles needs to be
balanced against other considerations: First, the definitions are part of a larger legal
framework and the implications of any application of legal theories must be carefully
considered from a policy perspective, not simply their usefulness in a particular sanctions
case. Second, the need for interpretation must be balanced against the need for legal
certainty, recalling again that the definitions of fraud and corruption are embedded in
legal agreements with the Banks member countries and harmonized with other MDBs.
Third, the fundamental principle of fairness embodied in the adage nulla poena sine lege
requires that any legal theories flow logically from a reasonable understanding of the
November 15, 2010

6

intended meaning and scope of the definitions, as evidenced by the text of the definitions
themselves or from their legislative history, not simply because of ex post convenience.

22. Notions of Natural Justice. Natural justice and considerations of fundamental
fairness play a threefold role in the Banks sanctions system. First, they inform the
formal substantive and procedural legal framework and can therefore be a source of
interpretation. Second, they can be called upon to resolve conflicts among sources of law
(see below). Third, as is the case in other judicial or quasi-judicial proceedings,
17
in cases
where no formal source of law provides an answer to a legal question, the EO or the
Sanctions Board mayon an exceptional basisdecide a matter ex aequo et bono
according to their best faith judgment of the demands of fundamental fairness.

23. Notions of natural justice, on the other hand, are not acceptable grounds for
overriding other sources of law, including the Banks own written law. While the
sanctions process must be grounded in fundamental principles of fairness and due
process, this is accomplished through their reflection in the written legal framework. If
any party believes that an aspect of that framework does not comply with such
fundamental principles, it is their right, and arguably their duty, to agitate for changes in
the framework. They may not, however, appeal directly to their views of natural justice in
order to unilaterally override the written law. It should be obvious that such an approach,
which must inevitably be based on the partys subjective opinion as to what constitutes
natural justice, is untenable.

24. National law. As LEG has advised previously,
18
national law is not binding on
the Bank, and it cannot be used to supersede the Banks own legal framework. However,
national law can provide a useful point of reference, both to establish the existence of a
general principle of law as described above, or simply as a possible approach to a
difficult legal issue for which the Banks own framework provides no clear answer.
However, we must be very cautious in applying national law concepts to sanctions cases
ex post; such applications generally amount to a retroactive application of norms.
National law concepts are more likely to be reference points for the prospective
amendments to the Banks legal framework, and, in fact, many of the proposals in the
current round of sanctions reforms are grounded in a survey of benchmark national
legal systems.

25. Sanctions Procedures. The above sources of law go to the substantive merits of
sanctions cases. The procedural aspects of cases are governed by the Sanctions Board
Statute, the Sanctions Procedures and related guidance materials that the Bank may issue
from time to time.
19
These materials go to issues such as the standard and burden of
proof, the weighing of evidence, and the like. As has been the view since Thornburgh,
20

LEG should provide guidance on the proper interpretation of the Sanctions Procedures,
even if in the past it has not been called upon consistently to resolve procedural issues.
21

26. The interpretation of the Sanctions Procedures, and the application of ad hoc
procedures in situations not covered by the formal Sanctions Procedures, relies on similar
sources of law, including the legislative history, authoritative interpretation and
November 15, 2010

7

occasionallyparticularly in order to fill procedural lacunae as in the case of recent
settlementsgeneral principles of law.

27. Conflicts among sources of law. Any conflict among sources of law should be
settled through the application of the foregoing hierarchy. Constitutional issues arising
under the Articles of Agreement, for example, supersede all other considerations.
Likewise, considerations of policy purposes may supersede the literal meaning of the
definitions of Sanctionable Practices, the Banks own legal framework supersedes
general principles of law, and so on.

28. A thornier issue arises when equivalent sources of law, such as different versions
of the legal framework, conflict. The most common such issue arises when the legal
agreement between the Bank and the borrower incorporates a certain version of the
definitions, as set out in a certain version of the Procurement or Consultant Guidelines,
and the bidding documents governing the procurement where the alleged Sanctionable
Practice occurred contain a different version of the definitions, or even no reference to
the Banks legal and policy framework for fraud and corruption at all.

29. The sanctions regime was developed as part of the Banks comprehensive effort
to ensure that loan proceeds are used for the purposes intended. The regime has its origins
in a 1996 decision of the Board of Executive Directors, which approved the introduction
of fraud and corruption provisions in the Banks Procurement and Consultant Guidelines
and the amendment of the General Conditions to provide contractual remedies for fraud
and corruption, as well as the establishment of the Sanctions Committee as predecessor to
the Sanctions Board. The definitions of Sanctionable Practices have undergone several
iterations since 1996, most recently in 2006, when the Bank adopted major reforms of the
sanctions regime. Among these was the adoption of new harmonized definitions for four
Sanctionable Practices (i.e., corrupt, fraudulent, collusive and coercive practices) among
the major MDBs and the inclusion (by the Bank only) of a new fifth Sanctionable
Practice (i.e., obstructive practice). In each case, the changes in definitions were adopted
prospectively, with no application intended to ongoing loans and projects.
22

30. Thus, the question arises: which version of the Guidelines, each of which may
contain different legal standards for defining the sorts of misconduct which may be
sanctioned by the Bank, should apply to a particular sanctions case? Is it the standard
generally in effect at the time of the conduct? Or the standard in the version of the
Guidelines incorporated in the legal agreement between the Bank and the borrower for
the relevant loan and project? Or, if different, as sometimes happens, the standard set out
in the bidding documents and/or the contract between the borrower and a contractor?

31. To analyze this issue correctly, we must start by distinguishing two different
issues which are often conflated: the establishment of jurisdiction over a case, on the one
hand, and the legal standards to be applied to the case on the other.

32. Jurisdiction. As discussed above, jurisdiction is established through the
application of any of the Procurement, Consultant or Anti-Corruption Guidelines that
November 15, 2010

8

include provisions establishing the Banks right to sanction to the project where the
Sanctionable Practice allegedly took place.
23
This application typically occurs through
the incorporation by reference of the relevant guidelines into the loan or other legal
agreement governing the project. This jurisdiction does not depend on, and is not affected
by, the provisions of any subsidiary agreement between the Borrower and third parties or
the bidding documents. Conversely, the absence of such provisions in the legal agreement
deprives the Bank of jurisdiction, notwithstanding the inclusion of such language
elsewhere.
24

33. The Bank does not need the agreement of third parties to sanction, since the right
to sanction is one that carries with it no corresponding obligation on the part of the
sanctioned party. For example, the Bank does not need a firms consent to issue a letter of
reprimand to that firm. In the case of debarment, there is arguably a corresponding
obligation, not on the part of the debarred party but rather on the part of borrowers, which
are required to respect the ineligibility of the debarred party in carrying out Bank
financed procurement. Even in cases where conditions are placed on non-debarment or
release from debarment, these are not contractual obligations but a unilateral
determination on the part of the Bank that it will either debar a party, or not release a
party from debarment, if those conditions are not met.
25

34. The agreement of third parties is necessary, on the other hand, to establish other
rights of the Bank that entail a corresponding obligation on the part of the third party. The
most common and important such right is the so-called third party audit right, which
permits the Bank to inspect the books and records of bidders, contractors and other
recipients of Bank financing and the corresponding obligation of such recipients to permit
the Bank access. Other such obligations are embedded in paragraph 10 of the Anti-
Corruption Guidelines, including the obligation to report allegations of Sanctionable
Practices and cooperate with Bank investigations.

35. It should also be mentioned that the inclusion of appropriate provisions regarding
Bank sanctions and fraud and corruption generally in bidding documents and contracts,
while not necessary to establish the Banks authority to sanction, serves two important
purposes relevant to the sanctions regime. First, as a matter of fundamental fairness, it
puts bidders and contractors on notice that they are subject to the sanctions regime. The
existence of secret rules offends fundamental principles of fairness and due process.
Second, these provisions bolster the Banks defenses against potential claims of tortious
interference with contract or defamation by sanctioned parties, for reasons examined in
depth in the July 2009 Audit Committee paper on Sanctions Reform.
26
The exclusion or
revision of these provisions without the Banks concurrence should therefore be viewed
as a serious performance breach by the borrower or project implementing entity.

36. The above is only a general discussion of the issues surrounding jurisdiction of
the sanctions process. LEG has provided detailed guidance on issues relating to subject
matter and in personam jurisdiction in the Sanctions Manual.

November 15, 2010

9

37. Legal Standards and the definitions of Sanctionable Practices. Once jurisdiction
is established, a separate question arises as to the applicable legal standards for judging
allegations of fraud and corruption. The first point to make is that, as mentioned above,
the legal standard is not necessarily the one most recently adopted at the time that the
conduct takes place. As explained above, the legal standards that apply to a particular
loan and project are established contractually as between the Bank and the borrower.
Under current practice, unless the legal agreement is amended, these standards apply for
the duration of the project.

38. A particular problem arises when, as not infrequently happens, the standards in
the legal agreement between the Bank and the borrower are at variance from those that
apply to a particular bidding process where a Sanctionable Practice is alleged to have
taken place, or those embedded in a contract financed by the Bank during the
implementation of which a Sanctionable Practice is alleged to have taken place. When the
Bank updates its procurement policies, borrowers are customarily given the option to
apply these new policiesincluding the corresponding updated standard bidding
documentsto bidding processes on a forward-looking basis. This should be reflected in
an amendment to the legal agreement, in which case the two are harmonized leaving no
ambiguity, but often the change is agreed informally between the TTL and line ministry
or project staff, not the borrowers designated representative, which is normally the
Ministry of Finance or equivalent.
27
Nevertheless, since the Guidelines themselves do
not specify the application of a particular version of the standard bidding documents but
only appropriate bidding documents, it can be argued that the Guidelines allow for
some flexibility for the Bank and borrower to agree on updated legal standards for
particular procurement, or for procurements generally, during project implementation.

39. Moreover, the Procurement Guidelines themselves state:

The rights and obligations of the Borrower and the providers of goods and works
for the project are governed by the bidding documents, and by the contracts
signed by the Borrower with the providers of goods and works, and not by these
Guidelines or the Loan Agreements.
28

40. Thus, the legal standards that govern the relationship between the borrower and
bidders and contractors are those set forth in the bidding documents and contract forms,
even if the legal agreement between the Bank and borrower contains different standards
through incorporation by reference of a prior or subsequent version of the Procurement
and Consultant Guidelines.

41. Although the Bank is neither party to nor bound by the bidding documents or the
contract between the borrower and the Respondent, and even if the Banks jurisdiction
does not depend on it, it is LEGs view that the Bank cannot simply ignore the fact that
the Respondent has been put on notice that it would be subject to sanction based on a
particular set of definitions, even in cases where those definitions are not the ones agreed
in the Loan Agreement. Often, the Bank itself has agreed to the application of updated
standard bidding documents that led to the mismatch in definitions, which means that all
November 15, 2010

10

three relevant partiesthe borrower, the contractor and the Bankall have reached the
same understanding on the applicable legal standards. In such cases, considerations of
equity should compel the Bank to accept the standards agreed as between the Borrower
and the Respondent as governing the particular contract and any sanctions case in
connection with the procurement or implementation of that contract. From a legal policy
point of view, this also often serves the interests of the Bank, since the definitions in the
bidding documents are more recent and therefore more comprehensive. However, this
also represents something of a double-edged sword: it also means that in those cases
where the bidding documents and contracts specify an earlier version of the definitions,
because, for example, of a lag between the adoption of new policies and the
implementation of new standard bidding documents, the Bank must be willing to accept
the application of earlier, less comprehensive definitions.

42. In cases where the bidding documents are silent on the issue, however, the above
considerations of equity do not, in our view, apply. There are, essentially, no standards to
override those set out in the legal agreement, so those standards should apply.

B. Rules of Evidence and Standards of Proof

43. Consistent with the administrative nature of the proceedings, the Sanctions
Procedures provide for an extremely permissive approach to evidentiary issues. Section
7.01 of the Sanctions Procedures provides as follows:
Any kind of evidence may form the basis of arguments presented in a sanctions
proceeding and conclusions reached by the Evaluation Officer or the Sanctions
Board. The Evaluation Officer and the Sanctions Board shall have discretion to
determine the relevance, materiality, weight, and sufficiency of all evidence
offered. Hearsay evidence or documentary evidence shall be given the weight
deemed appropriate by the Evaluation Officer or the Sanctions Board. Without
limiting the generality of the foregoing, the Evaluation Officer and the Sanctions
Board shall have the discretion to infer purpose, intent and/or knowledge on the
part of the Respondent, or any other party, from circumstantial evidence. Formal
rules of evidence shall not apply.
44. So, in short, INT (or the Respondent) may present any kind of evidence.
Therefore, best evidence
29
is clearly not a requirement, although where it is available,
we see little reason for INT not to present it. Where best evidence is not available,
Section 7.01 permits INT to present whatever circumstantial evidence it can marshal to
support an allegation or factual assertion. At the same time, it is entirely up to the EO and
the Sanctions Board, in their discretion, to evaluate and weigh the evidence, so they may
freely decide that circumstantial evidence proffered by INT is insufficient to support the
relevant allegation or assertion.
30
In exercising that discretion, the EO and the Sanctions
Board should consider all the relevant circumstances and the inferences that may
reasonably be drawn from them (see below), bearing in mind that the standard of proof
for sanctions proceedings is more likely than not.

November 15, 2010

11

45. INT has asked LEG to advise on our understanding of the meaning of more
likely than not. It should be borne in mind that the standard of proof is predicated on the
same basic considerations that underlay the omission of an explicit mens rea requirement
from most of the definitions, namely the administrative nature of the proceedings and
INTs lack of investigative tools. This standard of proof is understood as being equivalent
the preponderance of the evidence, essentially the standard to be found in civil cases in
most jurisdictions. When this nomenclature adopted in 2004, it was felt that the phrase
more likely than not would be more understandable to non-lawyers.
31

46. This history indicates that INTs view of the standard of proof as requiring more
than 50% of the corpus of evidence has some merit. Certainly, the standard is not meant
to be in any way equivalent to beyond a reasonable doubt that prevails in criminal law.
Therefore, the decision maker may still be left with reasonable doubts about the
culpability of the Respondent and nevertheless impose a sanction.

47. However, the more than 50% approach may not be the most useful or
appropriate either at the EO stage or at the Sanctions Board stage. Firstly, it suggests a
formulaic approach to the evidence while, as we saw in our analysis of Section 7.01
above, the Sanctions Procedures calls for a discretionary approach. Different pieces of
evidence may be given more or less weight, or judged more or less relevant or credible,
by the EO and the Sanctions Board in their discretion. While perhaps in theory this
process could be reduced to a mathematical formula, we do not view this as a feasible or
even desirable way to think about the exercise of discretion. Rather, the EO and the
Sanctions Board should look at the corpus of evidence before them as a whole and
determine whether the allegations are more likely than not to be true.

48. Moreover, the EOs assessment of the evidence requires special considerations,
given that he or she is looking at only one side of the case. The EO is actually charged
with assessing the sufficiency of the evidence, in contrast to the preponderance of the
evidence. The difference is explained in the Guidance Note for the IBRD/IDA Evaluation
Officer adopted in April 2009,
32
which reads, in relevant part, as follows:
Sufficiency of evidence differs from the preponderance of evidence required for the
imposition of sanctions by the Sanctions Board, since, at the time that the EO reviews
a proposed Notice of Temporary Suspension or Notice of Sanctions Proceedings, the
EO will not have seen any counter-evidence from the Respondent. The EO therefore
generally looks simply at the evidence presented in either type of Notice to
determine whether a reasonable person would conclude, considering all relevant
circumstances, that it was more likely than not that the Respondent had engaged in a
Sanctionable Practice.
33
When the EO has received the Respondents Explanation or
Preliminary Explanation, however, s/he may consider whether any counter-evidence
presented by the Respondent has cast enough doubt on the sufficiency of the
evidence to render it less likely than not that the Respondent has engaged in a
49. So, in laymans terms, the issue for the EO is: What would a reasonable person
make of the evidence before the EO, taking into consideration all of the known facts and
circumstances? And while it is not the EOs job to make the Respondents case, in
November 15, 2010

12

making this assessment, it is appropriate for the EO to consider other inferences that
might be drawn from the evidence beyond the inferences posited by INT. If an inference
that would exculpate the Respondent is at least as likely than an inference that would
inculpate it, then the EO may reasonably conclude that INT has not offered sufficient
evidence to show that it was more likely than not that the Respondent engaged in a

50. This is not to say that the EO should engage in idle speculation. It would be
unreasonable to require INT to disprove any and all hypothetical alternative explanations
for a Respondents actions. Any alternative inferences the EO considers should arise
logically and ineluctably from the record itselfas should any inferences posited by INT.
To take a facile example for purposes of illustration, the payment by a bidder to a
procurement official shortly before being awarded a bid should, on its face, provide
ample grounds for an inference that the payment was made to influence the officials
decision. INT need not exclude other possible hypothetical explanations unless there is
something specific in the record that suggests that alternative explanation. If, on the other
hand, the payment was made at a poker party, the EO would be justified in considering
whether the payment has another explanation before concluding that improper purpose
was the most likely.

51. While it can be argued that the EOs review of the evidence should be more
lenient than this, given the Respondents opportunity to make out its own before the
Sanctions Board, it is our view that, besides the clear intent of the legislative history,
there are strong legal policy reasons for a high level of scrutiny at the EO stage of
proceedings. Both as a matter of fundamental fairness and to protect the Bank against
unnecessary legal risk, the Bank should not impose the burden of litigation and the
economic cost of temporary suspension on Respondents without a robust vetting of INTs
arguments and evidence. Just as the EO should not engage in speculation in assessing the
evidence, INT should not pursue cases based on speculative evidence. Moreover,
overreliance on the adversarial process, as pointed out above, has certain drawbacks in
the context of the Banks sanctions regime. All things considered, we believe that the
method of considering the various inferences that may be reasonably drawn from the
record, within the parameters outlined above, provides the best assurance of a fair
outcome.

C. Mens Rea as an Element of Sanctionable Practices

52. Legislative History. The issue of mens rea is not a new one for the sanctions
regime. In the course of the 2006 round of sanctions reform, INT and LEG
34
collaborated
in successfully resisting calls by some EDs to include an explicit intent requirement in
the definitions of Sanctionable Practices. In so doing, INT and LEG relied on two
considerations: First, sanctions proceedings are not criminal in nature, and an intent
requirement is typical of criminal law, not civil or administrative law. Civil law liability,
for example, can arise from negligence or, where strict liability applies, no mens rea at
all. Second, INT lacks the investigative tools of a police force and should not be obligated
November 15, 2010

13

to demonstrate intent which, as it relates to an accused partys subjective state of mind, is
notoriously difficult to prove.

53. At the same time, it was recognized at the time that an implicit mens rea element
could be inferred from the definitions, arising from the requirement that misconduct be
undertaken in order to achieve a particular purpose and, typically, that said purpose be
improper.
35
This position is reflected in the Commentaries to the Anti-Corruption
Guidelines, which provides the following interpretation of mens rea in regard to corrupt
practices:
Unlike the definitions of corrupt practice found in most international
conventions, but like the Banks Procurement and Consultant Guidelines both
before and after the 2006 sanctions reforms, this definition does not include an
explicit and specific element of intent (i.e., evidence of the actual state of mind
of the accused party). Most international conventions require intent because they
are aimed at the criminalization of corruption, and mens rea is a fundamental
element of a criminal act. The Anti-Corruption Guidelines deal with contractual
obligations, on the one hand, and administrative measures which the Bank calls
sanctions, on the other... Further, because of the Banks ability to investigate
fraud and corruption in connection with loan proceeds is limited and does not, for
example, include the ability to subpoena witnesses or use other police powers, an
intent requirement would impose an unnecessarily high standard of proof.
However, the definition does require that the corrupt practice be undertaken for a
demonstrable purposeto influence improperly the actions of the target of the
corrupt practicewhich can be seen as an implicit element of intent. This
purpose can be shown either by direct evidence or, more typically, by reference
to a course of dealing, acts of the accused party or other circumstantial evidence
from which purpose can reasonably be inferred.

54. The very conscious objective of this approach was to avoid INT having to prove a
particular state of mind on the part of the Respondent, something thateven with the
flexible approach to evidence in the Sanctions Procedureswas thought to be likely to
prove an insurmountable obstacle in many cases.

55. Nevertheless, the sanctions regime was clearly not intended to penalize innocent
parties or conduct. Some changes to the definitions were made in the course of their
internal review at the Bank in order to ensure this was the case. Notably, the word
improperly was inserted in the definition of corrupt practice to ensure that it would
not inadvertently capture legitimate conduct. The Commentaries explain the requirement
as follows:
The definition includes an element of impropriety. This element may refer either
to the purpose of the corrupt transaction (e.g., to induce a breach of duty by a
public official) or to the means by which it is effected (e.g., a bribe) or to the
result (e.g., an undue advantage accruing to the corrupt actor). An element of
impropriety is necessary to the definition because it could otherwise be
interpreted to cover legitimate conduct (e.g., the payment of a salary to
influence an employee to perform his or her job). While the term improper
may be appear subjective on its face, it should be read in the light of international
November 15, 2010

14

conventions such as UNCAC, the OECD Convention and the Council of
Europes Criminal Law Convention on Corruption, which typically require that
corruption involve a breach of duty by the target of the corruption (see, e.g.,
UN Convention, Council of Europe Convention) and/or the receipt of an undue
advantage to, or the avoidance of an obligation by, that target (e.g., the OECD
Convention, UN Convention, Council of Europe Convention) or the party
seeking to influence the target (see, e.g., OECD Convention). The giving of a
bribe is therefore improper, whether or not the person accepting the bribe takes
any improper action, or even whether or not the bribe is intended to induce an
improper action (another reason why facilitation payments are not exempted). By
the same token, an otherwise proper payment or giving of a thing of value may
constitute corruption if it can be shown to have induced (or was meant to induce)
the target to breach a duty, for example a contribution to a charity favored by a
government official in order to secure the award of a contract to the contributor
that, under applicable rules, should have gone to another party.
56. It is interesting to note that, according to this interpretation, corrupt practice can
be found even in cases where there is no improper purpose, but where the means to
achieve the purpose is improper.

57. In the case of fraudulent practice, a specific element was added requiring that the
fraud be committed knowingly or recklessly. This issue, along with certain other issues
relating to fraudulent practice, is discussed separately and in detail below.

58. Mens Rea Requirements in National and International Law. If the Bank were to
read into these definitions a specific mens rea requirement, the question then arises, what
should that requirement be? The task is complicated by the fact that common and civil
law systems do not share exactly the same approaches to the issue. For example, the
common law term recklessness finds no easy counterpart in continental European
languages, while the civil law concept of dolus eventualis, while roughly equivalent to
recklessness, has no precise translation into English (or common law concepts).

59. The common law recognizes the following degrees of mens rea:

Specific intent: The offender has both the foresight (aka knowledge) to
understand that his actions may result in a particular result and the desire to
achieve that result, on a subjective basis.

Recklessness: The offender does not necessarily desire it, but is aware
36
of the
possible result of his actions and goes ahead anyway, thus exposing his victim
to the risk of harm.
37
The greater the probability of the risk, the more
egregious the recklessness and the greater the penalty imposed.

Negligence: The offender neither foresees nor desires the result of his actions,
but a reasonable person would have been able to foresee the result and
therefore refrained from the action. In criminal law, negligence is usually only
applied where the consequences of the offenders actions are so severee.g.,
November 15, 2010

15

negligent homicidethat society has an interest in punishing the offender
notwithstanding his or her lack of true intent.

Strict liability: The offender is held liable for his actions alone (actus reus)
regardless of his mental state. This standard is relatively rare but not unknown
in criminal law,
38
but is often applied in contract and tort law, for the purpose
of risk allocation.

60. The civil law tradition recognizes the following degrees of mens rea:

Dolus directus: The offender has both the foresight (aka knowledge) to
understand that his actions may result in a particular result and the desire to
achieve it, on a subjective basis. This therefore corresponds to common law
specific intent.

Dolus in the second degree: The offender is held culpable on dolus in the
second degree for collateral harm that is a necessary consequence of the
specific result that he or she desired and foresaw.
39
This can be seen as a form
of recklessness under common law.

Dolus eventualis: The offender is aware that, as a result of his behavior, an
offence could occur; the offender is aware of the possible result of his actions
and goes ahead anyway. This corresponds roughly to the common law concept
of recklessness.
40

Culpa: The offender neither foresees nor desires the result of his actions, but a
reasonable person would have been able to foresee the result and therefore
refrained from the action. Culpa in the civil law tradition implies a moral or
normative judgment with regard to an objective standard of care (allegedly
applied by a reasonable person). This corresponds to the common law concept
of negligence.

Strict liability. The concept of strict liability also exists under civil law. As
under common law, it is most often used in contract and tort law as a risk
allocation device.
41

61. While most international conventions on anti-corruption apply a specific intent
requirement, a requirement that the Bank consciously rejected, the Statute of the
International Criminal Court, which attempts to create a hybrid system melding both
common and civil law approaches, may be of interest to the Bank. Article 30 of the
Statute defines mens rea as requiring both intent and knowledge, with intent meaning
both (i) the intent to commit the act and (ii) either the desire or knowledge that a certain
consequence will ensue. Knowledge means awareness that a consequence will ensure in
the ordinary course of events.

November 15, 2010

16

62. Policy Implications of Mens Rea as an Element of Misconduct. If the Bank were
to specify the degree of mens rea required to establish culpability, it would need to
choose from among the foregoing options, bearing in mind the purpose of the sanctions
regime itself and the actual words of the definitions. Here we are faced with a quandary:
the words in order to which gave rise to this issue in the first place speak quite directly
to a desire to achieve certain results, which arguably suggests specific intent/dolus
directus, since neither recklessness nor negligence are predicated on the offender actually
have a specific purpose. So, if emphasis is to be put on mens rea, rather than the
transaction itself, it is hard to avoid adopting a specific intent/dolus directus requirement,
which we know from the legislative history the Bank did not intend. Bearing in mind the
legislative history, we should carefully examine the impact of a specific mens rea
requirement on INTs ability to bring and win sanctions cases.

63. It would also be appropriate to reflect on the implications for the range of
sanctions available under the current process and the relevant aggravating and mitigating
factors specified in the Sanctioning Guidelines. In most legal systems, the degree of mens
rea is not only a requirement for liability, but it may determine either the way in which
the offense is characterized (e.g., murder vs. negligent homide) and/or the punishment. At
the moment, the sanctions system takes mens rea into account in only a few instances, in
particular under the guidance on corporate groups, where failure to supervise (i.e.,
based on negligence) by a controlling person may give rise to a sanction under
Section 9.04 of the Sanctions Procedures, but specifies that the sanction would not
normally be debarment but rather, depending on the prevalence of the failure, either
conditional non-debarment or a letter of reprimand. Similarly the former Sanctioning
Guidelines recognized the egregiousness of the conduct as an aggravating factor, which
could be interpreted as providing for additional debarment time for higher degrees of
mens rea, although we are not aware of any actual case where the EO or the Sanctions
Board has invoked the factor in this way. The new Sanctioning Guidelines do not
recognize mens rea as an aggravating or mitigating factor, although it is arguably implicit
in some of the factors cited (e.g., acting as ring leader, use of sophisticated means).

64. LEGs Recommended Approach. In any event, these are questions of legal policy
that do not have a clear pre-ordained answer. It is our view that the best course to take in
regards to mens reaand the one closest to the spirit and intent of the 2006 reforms
would be to simply follow the strict letter of the definitions, without introducing a formal
mens rea requirement.

65. Given the safeguards already built into the definitions, we would question
whether a mens rea requirementat least one that is disaggregated as an element of the
definitions and argued as a separate matteris either advisable or necessary. If, for
example, it can be shown that money passed hands from a bidder to a government
procurement official in connection with a procurement, and it is clear from the
circumstance that the purpose of the transaction was to influence a procurement decision,
we see no significant value in requiring a showing that the bidder acted with specific
intent, recklessly or even (if such a thing is possible) negligently. Nor do we see a
significant risk that an innocent party would be sanctioned. One of the reasons for
November 15, 2010

17

avoiding an explicit mens rea requirement in the 2006 definitions (other than fraudulent
practice) was to take the focus off of the subjective state of mind of the Respondent and
on to more easily provable objective facts, the course of dealing from which the purpose
of the transactions under examination can be inferred. On the other hand, it is true that the
Respondents state of mind may equally be inferred from this same course of dealing
but, again, this is arguably merely an alternative way to conceptualize the essentially
same analysis.

66. Therefore, in our opinion, once the EO determines that INT has made out a prima
facie case that an act was undertaken for an improper purposea case that may be based
on reasonable inferences from objective facts, not necessarily the Respondents state of
mindit should then be up to the Respondent to attempt to rebut by showing that the
action did not, in fact, have an improper purpose: to follow on the example cited above,
that its payment to a procurement official was, in fact, repayment of a debt. Or that a
particular misrepresentation was due to mere error, not any intent to mislead. In this
sense, while mens rea would not be an element to establish culpability, the lack thereof
could be used by Respondents as an affirmative defense. As for any affirmative defense,
the burden of proof would be on the Respondent to make the appropriate showing.

D. Theories of Liability

67. Some of the most vexing issues for the disposition of sanctions cases are those
concerning the extent to which theories of joint or secondary liability may support the
sanctioning of persons who have not physically perpetrated the wrongdoing.

68. In this connection, it is important to set out some basic distinctions made in the
legal framework as to the basis on which a party may be sanctioned. The first is between
named Respondents, on the one hand, and other parties which may be sanctioned under
Section 9.04 (formerly Section 19(4)) of the Sanctions Procedures, on the other.
Respondents must be culpable, which is to say that they must be directly involved in the
actual Sanctionable Practice, while other parties may be held responsible, even though
they had no hand, either directly or indirectly, in the Sanctionable Practice.
42

69. It is also important to distinguish between the 2004 and 2006 definitions of
corrupt and coercive practices, which, as mentioned, include the words directly or
indirectly, on the one hand, and prior definitions, on the other. LEG has already
commented on the importance of this change in its 2008 advice in connection with the
NRIMP case.
43
The legislative history is clear that these definitions prior to 2004 did not
include aiding and abetting or other forms of secondary liability; the introduction of the
phrase directly or indirectly was a conscious effort to address this loophole in the
early definitions:
44

The first issue [whether the pre-2004 definition of corrupt practice includes
within its scope participatory acts such as aiding and abetting, complicity and/or
conspiracy] has a fairly straightforward answer. The record strongly suggests that
the 1999 definition of corrupt practice did not capture secondary forms of
wrongdoing such as aiding and abetting, and that the addition of the words
November 15, 2010

18

directly or indirectly to the definition in 2004, as proposed by Management
based on the Thornburgh reports and approved by the Executive Directors, was
intended to close this lacuna. The Presidents Memorandum to Executive
Directors covering the 2004 versions of the Procurement and Consultant
Guidelines specifically stated that the proposed changes to the definitions relating
to fraud and corruption were intended, among other things, to capture indirect
participation in corrupt activities.
45
The 2000 Thornburgh report had alluded to
the need to refine the 1999 definition of corrupt practice, and the 2002
Thornburgh report had found that the 1999 definitions needed to be
supplemented to assure coverage of, among other things, an action in
furtherance of an advanced conspiracy, fraud or corruption accomplished
through a middleman and acts constituting assistance to others in executing
fraud and corruption.
46
The 2002 report had recommended that these existing
gaps be corrected, and that incomplete forms of the offenses, and forms of the
offenses accomplished largely by others, be incorporated as well.
47,48

70. The Concept of Culpability. INT has asked for LEGs view of the proper
interpretation of the requirement that Respondents be directly involved in the
wrongdoing, bearing in mind that a narrow interpretation would limit Respondents to
those who participate personally in the physical act (e.g., the bag man for a bribe). We
should begin by saying that, as mentioned above, the direct involvement requirement
does not necessarily mean that others will not be liable. The 2004 and 2006 definitions of
corrupt practice allow for direct or indirect action, thus encompassing various forms of
indirect participation, including the use of agents, aiding and abetting, and the like. The
introduction of collusive practice captures various forms of indirect participation as
well. Moreover, Section 9.04 of the new Sanctions Procedures provides a basis for
sanctioning related persons, even where direct involvement in the Sanctionable Practice
cannot be established.

71. Nevertheless, we are of the view that culpability should extend beyond carrying
out the predicate physical act(s) that constitute the Sanctionable Practice. It would create
a perverse result if a party could escape culpability simply by using another as an
instrumentality to commit a sanctionable practice. Instruction or orders, as well as
plausibly lesser forms of involvement such as approval or guidance, should provide
grounds for culpability and therefore naming the party as a Respondent, even under pre-
2004 definitions. In such cases, long-standing theories of agency
49
or co-perpetration,
50

commonly understood across legal systems, provide a solid basis for holding such parties
culpable. On the other hand, mere knowledge of or non-intervention in a Sanctionable
Practice clearly should not grounds for culpability andbarring a duty to supervise such
as in the case of a parent company with respect to its subsidiaries or managers with
respect to their subordinatesshould not be grounds for sanction on the basis of
responsibility.

72. Respondeat Superior and Vicarious Liability. We support the notion that firms
should be sanctioned as Respondents for the acts of their employees for or on behalf of
the firm. Assuming that such acts are within the employees actual or apparent authority,
they are actions of the firm as a matter of law, so it is not entirely clear that even we need
November 15, 2010

19

resort to respondeat superior as a basis for liability. The practice of sanctioning of firms
for the acts of employees, to our knowledge, has been routinely followed in the
jurisprudence of both the Sanctions Committee and the Sanctions Board over the years.
51

73. The theory of respondeat superior may also be useful where the authority of the
employee is either in doubt or difficult to prove. From a legal policy point of view, we
would agree that requiring a showing that a particular employee was specifically
authorized to commit the Sanctionable Practice would place an unacceptable evidentiary
burden on INT, for the reasons we have already discussed. It should be sufficient that the
employee acted on behalf of the firm to establish a prima facie case that the firm would
then need to rebut through the interposition of an affirmative rogue employee defense.
52

74. If the firm can show that the person who engaged in the wrongdoing was a true
rogue employee (e.g., acting on his or her own, without the authorization or knowledge
of his supervisors and in violation of corporate policy), then this could be the basis for
either full exculpation or a significant mitigating factor in determining the appropriate
sanction for the firm. Where the firm can show that there was no failure to supervise and
that appropriate corrective measures were taken as soon as the wrongdoing come to light,
then that could be grounds for exculpation. The tripartite test proposed by INT strikes us
as a reasonable approach. In each case, this being an affirmative defense, the burden of
proof should rest on the firm.

75. Managers and other Controlling Persons. A different approach, in our view,
would be appropriate when considering the culpability of managers, supervisors and
other controlling individuals. In such cases, we would turn to concepts akin to those that
apply to the criminal liability of individuals, where a corporate officer is held liable for
acts performed by other corporate agents only when it can be shown that they acted under
his or her direction or with permission; mere knowledge or passive acquiescence in
wrongdoing by subordinates is not sufficient. However, given the special circumstances
in which the Banks sanctions regime operatesincluding the limitations on INTs
investigatory powers and the fact that the regime is administrative in natureit is our
view that evidentiary requirements may be relaxed in appropriate cases. In cases
involving small, closely held companies, for example, it would not be unreasonable to
infer authorization if the manager exercises close supervision over, and has intimate
knowledge of, all of the companys business operations.

76. Responsibility of managers and other controlling persons as a basis for sanction as
a non-Respondent (generally not debarment) may be established in the usual fashion
through a showing of a failure to supervise or similar default.

77. Joint Criminal Enterprise and other Theories Based on Conspiracy. Since
2004, participation in a joint scheme is itself sanctionable as collusive practice, although
the 2004 definition, as well as the 2006 definition as modified through footnotes in the
Procurement and Consultant Guidelines, are limited to bid rigging schemes. So the
issue goes not to whether such conduct should be sanctionable, since it is (at least under
the 2004 and 2006 definitions) but rather whether participation in a joint enterprise
November 15, 2010

20

should be a basis for sanctioning a party as a culpable principal in another predicate
offense such as corrupt or fraudulent practice. It is, in essence, a subset of the question
above as to what constitutes culpability.

78. To apply joint criminal enterprise or similar theories based on conspiracy as a
theory of liability in Bank sanctions cases, it must be shown that the theory has been
already established as a general principle of law at the relevant time. As INT has
correctly pointed out, joint criminal enterprise has been adopted by international criminal
tribunals. Our research indicates that it was largely created by the judges and prosecutors
of the Yugoslav Tribunal, its prominence dating from the case of Dusko Tadic in 1999.
53

The rationale behind the theory in that case was that holding persons who conspired to
commit war crimes liable merely as aiders and abettors would understate their role and
degree of responsibility. The theory allows them to be held liable as principals. The other
rationale was the horrendous nature of the war crimes involved, including genocide,
under wartime conditions that made it difficult to prove the culpability of individual
perpetrators. The prospect of such crimes going unpunished or under-punished because of
lack of direct evidence of individual culpability seemed, understandably, unacceptable.

79. These rationales do not seem particularly apposite to the typical sanctions case.
However, concepts of joint liability akin to joint criminal enterprise as used in the Tadic
case have existed across legal systems for over a century.
54
The core elements of joint
criminal enterprise include: (1) a plurality of persons, (2) existence of a common plan and
(3) participation of the accused in the common plan. The plan itself may be established
by direct or circumstantial evidence. Participation, in most jurisdictions, must be
established through the showing that the accused committed some overt act in furtherance
of the common plan; passive acquiescence or mere knowledge of the plan is not sufficient
to establish participation.
55

80. International customary law recognizes three forms of joint criminal enterprise
56

two of which are potentially relevant to the Banks sanctions regime:

Basic: Under the basic form of joint criminal enterprise, each participant in the
joint enterprise is culpable for acts committed that a part of the common plan.

Extended: Under the extended form of joint criminal enterprise, participants may
also be liable for acts other than those which were specifically included in the
common plan if (i) it was foreseeable that the act might be perpetrated by one or
other participants in the joint enterprise, and (ii) the accused was aware that the
act was a possible consequence of the execution of the plan, and with that
awareness, the accused decided to participate in the enterprise.
57

81. While the exact parameters of any theory of liability drawing on joint criminal
enterprise should be worked out, on a case by case basis, through the development of
jurisprudence, it is our view that the extended form of joint criminal enterprise is not
consistent with the Banks legal framework for sanctions, as it relies on mere
foreseeability and therefore essentially sets a simple negligence standard for culpability.
November 15, 2010

21

However, we would support a concept that extends the basic form of joint criminal
enterprise to allow for culpability of acts that the Respondent either knew or, under the
circumstances, must have known were likely to occur as a result of the enterprisea
standard similar in nature to the scienter required for recklessness in fraudulent practice,
as explained below. In any event, the common plan required for establishing a joint
criminal enterprise will no doubt need to be established by inference from circumstantial
facts in many if not most cases, so the difference between the two standards is likely to be
largely academic, and the question of what acts lie within or without a common plan, and
what collateral acts should be considered a likely consequence of the plan, will need to
be argued by INT, and determined by the EO and Sanctions Board, on a case by case
basis, based on the particular circumstances of the case.

82. In our view, given its long-standing wide acceptance across legal systems,
theories of liability based on notions akin to joint criminal enterprise may be considered
as constituting a general principle of law and therefore a legitimate source of law for the
Banks sanctions regime. However, it is also our view, in light of the legislative history,
that application of this theory to pre-2004 definitions would constitute a retroactive
application of norms. As quoted above, the words directly or indirectly were introduced
into the 2004 definitions specifically to remedy certain lacunae in the pre-2004
definitions, including actions in furtherance of an advanced conspiracy. The same logic,
then, that compelled us to take the view that the pre-2004 definitions did not encompass
aiding and abetting and other forms of secondary liability compel us equally to conclude
that those definitions do not encompass joint liability beyond situations of true agency or
co-perpetration.

83. Accessories. As explained above, the words directly or indirectly were
introduced into the 2004 and 2006 definitions of corrupt and coercive practices for the
purpose of capturing aiding and abetting, and other forms of secondary liability. It is
therefore our view that accessories may be sanctioned as culpable parties (i.e., as
Respondents) under those definitions. The distinction between accessories and those who
may be held liable as joint principles under joint criminal enterprise and other conspiracy
theories lies in their motivation: whether the Respondent wishes the perpetrate the
offence as his or her own (animus auctoris) or merely wishes to support someone elses
actions (animus socii).
58

84. Traditionally, accessories were not punished with the same severity as principals,
although the common law distinctions between principals and accessories have largely
been abolished. The new Sanctioning Guidelines do not make a distinction between
principals and accessories, per se, but they do establish playing a minor or marginal
role as a mitigating factor, while acting a ring leader is an aggravating factor.

85. As stated in the 2008 Legal Memorandum, the legislative history shows that the
pre-2004 definitions did not encompass aiding and abetting or other secondary forms of
liability. As for the lack of conspiracy-based sanctionable practices, this is a legacy issue
that the Bank must live with.
November 15, 2010

22

!!/ !""#$" *$,+0!)( 0& '*+#1#,$)0 2*+.0!.$

86. LEG has also been asked to opine on several issues specifically relating to
fraudulent practice, which has, by far, formed the most common grounds for recent
sanctions cases. These issues include (i) the proper understanding of the term
recklessness as used in the 2006 definition of fraudulent practice; (ii) whether the
knowing and reckless standard should also be applied to cases based on pre-2006
definitions; and (iii) whether an omission may be considered a misrepresentation of fact
for purposes of the pre-2004 definitions of fraudulent practice.

A. Recklessness and Fraudulent Practice

87. The World Bank Group currently defines fraudulent practice as follows:
Fraudulent practice is any act or omission, including a misrepresentation, that
knowingly or recklessly misleads, or attempts to mislead, a party to obtain a financial or
other benefit or to avoid an obligation. (emphasis supplied)
88. A footnote to the definition in the IBRD/IDA Anti-Corruption Guidelines
explains that:
[t]o act knowingly or recklessly, the fraudulent actor must either know that the
information or impression being conveyed is false, or be recklessly indifferent as to
whether it is true or false. Mere inaccuracy in such information or impression, committed
through simple negligence, is not enough to constitute fraudulent practice.
89. This definition is unique among the harmonized definitions of Sanctionable
Practices in that it contains an express mens rea
59
element, knowingly or recklessly.
This element was introduced in the course of discussions in the sanctions reform working
group in 2006, principally to address concerns expressed by IFC and MIGA
representatives that misrepresentation through negligence should not be sanctionable. All
concerned agreed that the essence of fraud did not encompass a mere mistake,
consistent with the overall purpose of the sanctions regime to enforce ethical, not simply
professional, standards.
60
For the same reason, it has been agreed among INT and LEG
that this standard should also be read into earlier versions of the definition, even though
they did not contain this express element (see below).

90. In many cases, the knowingly or recklessly standard has not proven problematic.
When a firm or consultant submits a bid, for example, that overstates its/his/her
experience, it is usually a straightforward matter to conclude that the misstatement
constitutes a fraudulent misrepresentation since the false information being conveyed is
within the direct personal or corporate knowledge of the accused. And firms have
routinely been found liable for the misrepresentations by their employees or agents under
the theory of respondeat superior.
61

November 15, 2010

23

91. More difficult issues arise, however, when a firm or individual relies on and
conveys information supplied by another. For the sake of convenience, we will call this
type of case indirect fraud. By way of example, where the chief executive officer of a
firm signs a bid (often several hundred pages long) that contains one or more
misrepresentations, should he or she be personally sanctionable in the absence of proof of
his/her direct knowledge of the inaccuracy of the information or even that the information
has been included in the bid? Or where a would-be primary contractor includes false or
misleading information provided by a subcontractor in its bid, should the primary
contractor be liable for the misrepresentation? Where there is evidence that the CEO or
primary contractor had actual knowledge of the misrepresentation in question, there is no
issue, but that is not usually the case.

92. The key issue in such cases, then, is the exact meaning of recklessness as used
in this definition. The explanatory footnote provides some guidance, but not enough to
settle some hard cases. LEG has therefore been asked to provide further guidance. In
considering this issue, LEG has reviewed the two principle general concepts of
recklessness, one under criminal law and the other under tort law, as well as a number of
relevant national laws in which recklessness has been applied to civil fraud.

93. Criminal Recklessness. As cited above, the contemporary common law concept
of criminal recklessness under common law requires awareness of a risk of harm, of
which the defendant is aware and proceeds anyway, thereby causing the harm to occur. It
is commonly said that recklessness requires indifference to, acceptance of or resignation
to a known risk. This notion of reckless indifference is reflected in the explanatory
footnote in the Anti-Corruption Guidelines cited above.

94. Criminal recklessness requires the accused to be aware of a risk that a certain
consequence may ensue from their action that they choose to consciously disregard. It is
therefore dependent on the subjective state of mind of the accused. Recklessness in this
sense is sometimes described as having a devil may care or dont give a damn attitude.
In the context of fraudulent practice, this would mean that the accused knows that the
information being conveyed (for example, in a bid) may be false or misleading, but
nevertheless goes ahead without checking its accuracy.

95. Under common law, the risk involved must be material, in both senses of the
word. First, the risk must be not merely remote or theoretical, but substantial and
unjustifiable.
62
Second, there must be a reasonably close nexus between the risk and the
situation in which the accused found themselves. Common examples include running a
red light (where the red light makes the accused aware of an immediate, substantial and
unjustifiable risk of a crash), carrying a gun during a fight (where the risk that the gun
might go off is obvious)
63
or hitting a child (where the risk of serious injury to the child is
equally clear).
64

96. This concept of recklessness is distinguishable from negligence, or even gross
negligence, by its subjective nature, i.e., its requirement of a showing of awareness on the
part of the accused. By contrast, negligence applies an objective test: a reasonable
November 15, 2010

24

person should have known that the information could be inaccurate. Negligence thus
imposes a duty of care to check on the accuracy of information being conveyed, without
the need to show that the accused had knowledge of the inaccuracy or a material risk of
inaccuracy.
65

97. Tortious Recklessness. The other principal way to understand the term
recklessness is grounded in tort law, as well as more traditional criminal law
approaches. This approach finds some support in the drafting of the Banks legal
framework, in particular the IBRD/IDA Anti-Corruption Guidelines. It should be noted
that the explanatory footnote to the definition of fraudulent practice specifically excludes
simple negligence, i.e., the failure of an actor to abide by the standards of a reasonable
person under the same circumstances. The modifier simple is generally used in contrast
to gross negligence, which is understood as a particularly egregious deviation from the
due care that a reasonable person would have applied in the circumstances, taking a
substantial and unjustifiable risk.

98. Under this concept of recklessness, simple negligence, gross negligence and
recklessness lie on a continuum: each involves unjustifiable risk-taking, differing only in
the degree of the actors deviation from the standard of due care that a reasonable person
would have applied in the same circumstances (i.e., when faced with the risk in question).
Tortious recklessness may be seen as parallel in concept to criminal recklessness, in that
the wrongdoer must have taken a substantial and unjustifiable risk, but is distinguished
from criminal recklessness in that the Respondents subjective awareness of that risk
need not be shown.

99. Recklessness as Applied to Civil Fraud under National Law. We have also
examined three areas of US national law in which the concept of recklessness has been
applied, either legislatively or by judicial interpretation, as an element of fraud: (1) Rule
10b-5 of the Securities Exchange Act, (2) the books and records provisions of the
Foreign Corrupt Practices Act (FCPA) and (3) the False Claims Act.

Rule 10b-5: Under this rule, the Securities Exchange Act
66
provides a civil cause
of action for persons who claim to have been defrauded by the issuers of
securities. While the Rule itself does not specify a mens rea element, over the
years since its adoption, courts have interpreted the rule to include a knowing or
reckless standard.
67
Although recklessness is subject to varying interpretations,
most courts have taken the view that recklessness under Rule 10b-5 requires a
showing that the defendant acted without regard to the truth of the statement. The
standard, as stated in one appellate case, is as follows:

Reckless conduct may be defined as a highly unreasonable [act or] omission,
involving not merely simple, but even inexcusable negligence, but an extreme
departure from the standards of ordinary care, and which presents a danger of
misleading buyers or sellers that is either known to the defendant or is so obvious
that the actor must have been aware of it. [emphasis supplied]

November 15, 2010

25

FCPA: The books and records provisions of the FCPA
68
place an obligation on
publicly traded companies to maintain proper records and maintain an internal
system of controls to properly monitor the record-keeping obligation; issuers
may be liable for knowingly falsifying their books, records or accounts. The
courts have held that this scienter requirement was introduced to codify SEC
enforcement policy that penalties not be imposed for insignificant or technical
infractions or inadvertent conduct, but not to provide a defense for those who
shield themselves from the facts.
69

False Claims Act: This statute
70
holds parties liable for knowingly presenting a
false claim for payment or approval to the US government. The term knowingly
is defined to include willful blindness or reckless disregard of the truth or falsity
of the information. Courts have interpreted this language to include, inter alia,
failures to explore a credible concern about billing or other records,
71
seriously
deficient record-keeping systems or failures to inquire that rise to the level of
deliberate ignorance.
72
At least one court has found that the Act was intended to
place at least a limited duty of inquiry on the defendant in order to reach what
has become known as the ostrich type situations where an individual has
failed to make simple inquiries that would alert him that false claims are being
submitted.
73
Some courts have also found that the Act encompasses extreme
carelessness.
74

LEGs recommended approach
100. Looking at these various approaches holistically, it is clear that there are some
core elements of the concept of recklessness common to each approach, across legal
systems and across substantive areas of law: the existence of a substantial and justifiable
riskin the case of fraud, the risk that the information being conveyed is either false or
misleadingcoupled with either actual awareness of the risk, willful blindness to the
risk, or a kind of cluelessness as to that risk. The main difference among approaches
in particular what sets criminal recklessness apart from other approachesis the
insistence of specific evidence of actual, subjective knowledge of the risk.

101. It is our view that an approach to liability for fraudulent practice that adopted the
criminal recklessness requirement of a showing of actual, subjective knowledge on the
part of a Respondent in all cases would be inconsistent with the legislative intent of the
2006 reforms to avoid standards that would create unreasonable evidentiary burdens for
the Bank, bearing in mind as well, as was often stated when the 2006 definitions were
discussed with Executive Directors, that sanctions proceedings are administrative and not
criminal in nature. We would therefore propose a hybrid approach, focused on the core
elements of recklessness that may be divined from all the approaches we have examined,
while taking a permissive stance with regard to the element of knowledge.

102. This approach does require the exercise of considerable discretion and therefore
some basic guidance to assist the EO and the Sanctions Board in coming to their
determinations would be appropriate. We would propose an analytic framework,
November 15, 2010

26

described in the following paragraphs, for determining whether or not a Respondents
deviation from due care is sufficiently egregious so as to establish that the Respondent
took an substantial and unjustifiable risk that merits sanction, bearing in mind that
recklessness requires a deviation more egregious than either simple and even ordinary
gross negligence.

103. Cases based on actual or inferred knowledge of risk. As a threshold matter, if a
case can be made out following the classic criminal approachthat the Respondent knew
of a substantial and unjustifiable risk that a statement was false or misleading and ignored
itthat would, ipso facto, constitute recklessness. It is also to be preferred where
possible, for while such a case is undeniably harder to prove, once proveneven through
the drawing of reasonable inferencesit should be more straightforward to litigate,
revolving as it does around a tangible and unambiguous scienter standard.

104. In this connection, we recall that the Sanctions Procedures contain flexible rules
of evidence, which among other things allows the inference of knowledge from
circumstantial evidence. (See the discussion of evidentiary matters in Section I.B above.)
This flexibility was established, once again, in recognition of the difficulties of proving
fraud and corruption cases in the Bank Group context. INT is free to provide
circumstantial evidence that an accused party was aware of a risk, for example through
the existence of red flags, that put the accused on notice that a particular piece of
information may be false or misleading, from which it may be inferred that the accused
had knowledge of the relevant risk. Whether or not such an inference is reasonable to
draw from a particular set of circumstances is left up to INT to argue and for the EO, and
then the Sanctions Board, to evaluate.
75

105. Cases based on extreme carelessness amounting to recklessness. Even with such
flexibility, it may not always be possible to make out a case based on actual or inferred
knowledge of the risks involved. In these cases, we would recommend a concept of
recklessness that draws on general tort law concepts of extreme carelessness, as
evidenced by the sort of blindness (willful or otherwise) to obvious risks articulated in
civil fraud cases.

106. The analysis begins with a proper understanding of due care. Even if
recklessness should represent a particularly egregious deviation from this norm, due
care still provides the baseline against which a Respondents conduct should be
judged. Due care is usually conceived as the degree of care that the proverbial
reasonable person would exercise under the particular circumstances of the case. Put
another way, this standard asks what precautions a reasonable person would takein the
case of fraud, what steps a bidder should take to ascertain the accuracy of information
contained in the bidin light of a substantial and unjustifiable risk that at least some of
that information may be false or misleading.

107. The conduct of the Respondent should be judged against the nature of the risk, in
terms of both the likelihood
76
that the information being conveyed is false or misleading,
and the severity of the potential harm flowing from the communication of such
November 15, 2010

27

information. The higher either is, the higher the required duty of care. Conversely, the
lower either the likelihood or the severity of the potential harm, the less precaution is
required. There are at least two ways of viewing the requisite harm. First, the degree of
harm to the integrity of the procurement process, as measured by how material the
information is to the overall bid, will influence the requisite duty of care. Clearly the
Bank has the right to expect a higher duty of care as to ascertaining facts that are key to
establishing bid validity. Second, the consequences of the harm are also relevant: where
false or misleading information may result in the loss of life or dire economic
consequence, for example, the duty of care is naturally higher.

108. Against the nature of the risk, the benefit of the Respondents chosen course of
action, together the costs of taking further precautions are also taken into account. There
may be cases where a particular risk is justifiable in light of the potential benefits to be
gained by the Respondents chosen course of action. By benefit we do not mean to
include, of course, any prospective illicit gains to the Respondent, but rather the benefit
to the procurement process or the project concerned, in terms, for example, of efficiency
or cost savings. Similarly, taking a particular risk may be justifiable in light of the
exorbitant cost of guarding against it.

109. The proverbial reasonable person is usually conceived of in light of the
standards of a particular society. A typical definition
77
runs as follows:
reasonable person. 1. A hypothetical person used as a legal standard, esp. to
determine whether someone acted with negligence; specif., a person who
exercises the degree of attention, knowledge, intelligence, and judgment that
society requires of its members for the protection of their own and of others'
interests. The reasonable person acts sensibly, does things without serious
delay, and takes proper but not excessive precautions. Also termed reasonable
man; prudent person; ordinarily prudent person; reasonably prudent person;
highly prudent personThe reasonable man connotes a person whose notions
and standards of behaviour and responsibility correspond with those generally
obtained among ordinary people in our society at the present time, who seldom
allows his emotions to overbear his reason and whose habits are moderate and
whose disposition is equable. He is not necessarily the same as the average man
a term which implies an amalgamation of counter-balancing extremes. R.F.V.
Heuston, Salmond on the Law of Torts 56 (17th ed. 1977).
110. At common law, it was traditionally common to refer to the reasonable person
as the man on the Clapham omnibus. In the international and multi-cultural context in
which the sanctions regime operates, this standard obviously poses some challenges. At
the same time, it would be impractical (and wrong from a policy point of view) to try to
judge Respondents by different standards according to their nationality or location.
78

111. Ideally, such a standard would be derived from, if not actually stipulated by, the
Banks procurement policies, as articulated in the Procurement and Consultant
Guidelines and the related standard bidding documents. We understand, however, that the
Banks policies do not provide for a specific standard of care for the preparation of bids,
November 15, 2010

28

nor are there norms from which such a standard could be reasonably inferred, at least
none clear enough to provide a basis for a claim of fraud.
79
LEG would therefore advise
taking the following alternative approaches to the development of a Bank-specific
concept of an appropriate reasonable standard of care in the context of Bank operations:
First, if the bidding documents themselves specify a standard of care, that
standard may be used to judge the conduct in question. Where, for example, the
Respondent affirms the accuracy of the information contained in its bid, there is
no reason not to hold the Respondent so accountable.
Second, in appropriate cases, reference may be made to relevant industry
standards, which in many industries are internationally recognized, or customary
or firm-specific business policies, procedures or practices. The policies,
procedures or practices of the Respondent firm may be relevant, since a person
who is charged by a firms own policies, procedures or practices with the
accuracy of a bid may reasonably be held accountable if the bid turns out to be
misleading. In any case made out on these grounds, INT should present evidence
of the relevant standards, policies, procedures or practices that serve as the
reference point for due care.
Third, for cases where the points of reference outlined above are absent, we
would propose that standards be articulated over time through jurisprudence. The
EO and/or Sanctions Board will need to exercise discretion in deciding such cases
based on the individual facts of each case, taking a kind of common law
approach to the development of Bank law in this area. This will require them to
set out their thinking underlying their determinations as precedent for future
cases.
112. Once this threshold is crossed, the analysis would then turn to whether the
Respondents conduct has deviated so far from the notional due care in ascertaining the
accuracy of information which a reasonable person would exercise that it constitutes
recklessness.

113. Guideposts for analyzing recklessness: While there is no escaping the exercise
of considerable discretion, based on the above considerations, LEG would offer the
following guideposts for assessing claims of recklessness in fraudulent practice cases:

As explained above, the threshold question in each case of indirect fraud is:
Was the Respondent faced with a substantial and unjustifiable risk that the
information on which he/she/it relied was false or misleading? To be substantial
and unjustifiable, the risk must have evidenced both a significant likelihood that
the information being relied on is false or misleading, and significant potential
harm flowing therefrom. The risk should also have presented a clear and direct
nexus to the particular circumstances faced by the Respondent. Recklessness
cannot be grounded in theoretical, remote or generic risks.

November 15, 2010

29

As for the Respondents awareness of the risk, while a subjective awareness is
not required, the Respondent must be shown to have ignored significant and
obvious red flags that would have put any reasonable person on notice of a
substantial and unjustifiable risk that a piece of information on which the
Respondent is relying may be false or misleading. (This analysis would not be
very different, in practice, from the must have known standard applied in civil
fraud cases.)

Cases in which the Respondent took affirmative steps to avoid knowing about the
risk involved, known as willful blindness, also provide a basis for recklessness.
Such cases can be viewed either as a form of inferred or constructive knowledge
or as a form of extreme carelessness.

As pointed out above, the deviation from ordinary due care must be egregious,
extreme or wanton, without necessarily being deliberate. It is not enough to show
merely that the reasonable person would have acted differently in light of the
risks involved, but that the Respondents conduct represents a shocking
indifference to those risks. Courts have seen this standard as showing an utter
lack of care, a failure to exercise even that degree of care that a careless or
inattentive person would exercise.
80

AmJur2d usefully describes recklessness as follows: Conduct may be reckless
even though the actor does not recognize it as being extremely dangerous. It is
enough that he knows or has reason to know of circumstances that would bring
home to the realization of the ordinary, reasonable person the highly dangerous
character of his conduct. The inability to realize the danger may be due to the
actors own reckless temperament, or to the abnormally favorable results of
previous conduct of the same sort. The actor must know or have reason to know,
the facts that create the risk.
81

A standard akin to criminal negligence could also amount to recklessness: cases
where the Respondents actions or their consequences are so egregious that it
would undermine the purposes of the sanctions system, in particular general
deterrence, if his/her/its actions were not sanctioned by the Bank. As for any
case, the conduct or its consequences must be so egregious as to constitute an
ethical, and not solely a professional failure.

114. We would emphasize that these guideposts are not meant to be exclusive, nor
are they meant to be cumulative. In some ways, they present alternative ways to
conceptualize recklessness. A mentioned above, in our view the only way to develop a
true legal test for recklessness in the context of Bank sanctions cases is through the
development over time of Bank-specific jurisprudence.

115. It is beyond the scope of this Advisory Opinion (and LEGs role in the sanctions
process) to prescribe outcomes from the application of these guideposts to particular
fact patterns. It is clear, however, that certain situations lie within or without the
November 15, 2010

30

parameters of recklessness as we have defined it here. The mere act of signing a bid
document, for example, without additional facts or circumstances, is not enough to
establish personal liability fraudulent practice on the part of the signatory, even if the bid
contains false or misleading information. Nor is the inclusion by a bidder of false or
misleading information provided by a sub-contractor enough, in and of itself, to establish
that the bidder engaged in fraudulent practice.

116. Other Relevant Considerations. Two other considerations are relevant to cases of
indirect fraud under existing features of the sanctions process regardless of the standard
against which recklessness is judged:

117. Responsibility of Controlling Parties. The Sanctions Procedures and related
guidance
82
already provide for derivative liability for controlling persons under what is
essentially a negligence standard. Persons who control a Respondent may be sanctioned
if it can be shown that their failure to supervise enabled the misconduct of the
Respondent. It is stated that these controlling parties are held responsible for the
misconduct, even if they are not culpable as wrongdoers. This provision is most often
invoked in cases of parents and subsidiaries, but could be applied more broadly. Thus a
CEO may be held responsible for a fraudulent bid by a firm for having failed to put into
place proper procedures to ensure the accuracy of information contained in their bids.

118. Scienter and the Choice of an Appropriate Sanction. It should also be noted that
the degree of severity of the Respondents conduct is a factor relevant to the choice of an
appropriate sanction.
83
While a Respondents actions may meet the threshold for
recklessness, the appropriate sanction for fraudulent practice based on recklessness would
not normally be as severe as the sanction for fraud based on actual (or inferred)
knowledge. Moreover, the guidance on corporate groups specifies that derivative liability
based on a failure to supervise does not normally lead to debarment, but to lesser forms
of sanction such as reprimand or conditional non-debarment.

119. Procedural Implications. In order to implement this approach, we would advise
the EO and the Sanctions Board to exercise discretion, based on the analytical framework
outlined above, in determining whether a particular conduct of a Respondent is reckless,
setting out clearly their thinking underlying their determinations so that specific standards
may be developed over time.

120. We recognize that this approach represents a departure, particularly for the Bank
EO, in that it requires that the Respondents conduct be judged against a highly
discretionary standard. The Bank EO will be required to go beyond its usual role of
simply applying the law and play a part in making the law. In this regard, without
involving LEG in the merits of particular cases, the Bank EO may wish to request LEG
advice on the application of this standard in cases of first impression.

121. Furthermore, given the prominent role that jurisprudence will play in the
development of a Bank-specific standard for recklessness, there is a strong argument that
a full adjudication of the facts, which is only available at the Sanctions Board stage of
November 15, 2010

31

proceedings, should be encouraged, in particular at the earlier stages in the development
of this standard. We would therefore recommend that, in exercising its discretion, the EO
give due consideration to colorable arguments posited by INT in cases of first impression
and, in cases of doubt, to allow such arguments to reach the Sanctions Board. At the same
time, we would stress the need for all actors in the process to respect the Bank EOs
autonomy of decision-making and the inviolability of her determinations once made.

B. Application of the Knowing and Reckless Standard to pre-2006 Definitions

122. As explained above, the inclusion of the knowing and reckless standard was
understood at the time as a clarification, not a limitation on the scope of fraudulent
practice. It is therefore our view that that standard may be seen as an implicit element of
pre-2006 version of the definition and be used as a benchmark for the mens rea required
to make out a sanctions case based on those definitions of fraudulent practice.

C. Omission as an Element of Fraudulent Practice in pre-2004 Definitions

123. Omission was first mentioned as an explicit element of fraudulent practice in the
2004 definition of fraudulent practice. However, it is generally understood across legal
systems that fraud may be committed through the omission of a material fact as well as
non-verbal acts.
84
The conduct that may constitute fraud relates not to the particular
modality of its perpetration but in its tendency to mislead.

124. The Banks own legislative history, as it relate to the definition of fraudulent
practice as it has evolved over time, contains no suggestion that the pre-2004 definitions
were intended to exclude omission.
85
It is therefore our considered view that the inclusion
of 'omission' in the 2004 definition and the formulation of 'acts or omission, including a
misrepresentation' in the 2006 definition were both intended as clarifications of existing
understandings of the sanctionable practice, not expansions of their scope.

D. Establishment of Prospective Fraud in Bids

125. Some recent sanctions cases have presented a difficult issue regarding possible
bait and switch schemes in connection with consultancy contracts, where the
Respondent wins a bid based on a particular named team of consultants, some or most of
whom do not eventually form part of the team working on the assignment. From the
circumstances of these cases, it seems possible that the Respondents never in fact
intended to use the consultants named in their bids. Proving this, however, can be
challenging, and LEG has been asked for its view of the appropriate standards to be
applied.

126. Clearly, the analysis must begin with the establishing whether or not a
misrepresentation occurred. The mere fact that the bid named consultants who were
subsequently changed is not, in and of itself, enough to establish that at the time of the
bid, the Respondent did not intend to use the original consultants. Here we immediately
encounter a threshold challenge, as the Banks Consultant Guidelines and standard
November 15, 2010

32

bidding documents do not provide a definitive answer as to what a bidder is actually
representing when it submits a list of consultants, at least for purposes of establishing
fraud. There may be cases in which the bidding documents themselves provide an
answer,
86
but even where that is not the case, it seems intuitively obvious that at least
some assurance must be implicit in a bid that includes a list of named consultants, or else
bids would be essentially meaningless. At the same time, we understand that under the
business model for consulting prevalent todayone in which consultancy firms rely on a
market of independent experts rather than their own permanent staff to meet bidding
requirementsit is not uncommon that changes in consultancy teams take place after a
bid is submitted. Consultancy contracts typically allow for changes in the consultancy
team so long as the new consultant is of equivalent qualifications, the fee does not
increase, and the change is agreed by the client.

127. Despite this ambiguity, there are at least some facts under which we believe that a
colorable case of fraudulent practice could be made, in particular what the bidder has
actually done, or not done, to confirm the availability of the named bidders. It seems to us
obvious that the evidence must show whether or not the named consultants were
available to the bidder or not at the time of the bid, which can be most directly and
convincingly established through confirmation by the consultants themselves. The typical
fact-patterns will lie along a continuum of possible arrangements: at one extreme, if the
bidder had no prior contacts with a particular consultant and made no efforts to contact
him or her in advance of the bid, we would argue that a strong inference of fraudulent
intent may be drawn. Moreover, a repeated pattern over time of the same bait and
switch by the same firm over several bids would, in our view, be highly suggestive. And
where a bidder has a general consent from the consultant to use his or her name in bids,
but made no effort to confirm the consultants availability for a particular assignment, we
would be of the view that this conduct would tend to show that the Respondent was
reckless in ascertaining his or her availability, But less clear-cut cases may easily occur,
and each case should be considered on its own merits, in the light of the requirements
discussed above.

III. &03$* !""#$"

A. The Sanctioning of Government Officials.

128. It has long been the Banks policy that governments and government officials
should not be sanctioned, based on the Banks structure as a cooperative institution in
which most of its borrowers are also member governments or subnational governments. It
is also grounded in respect for the sovereign status of its members. This policy of
immunity for government and government officials is embedded expressly in the Anti-
Corruption Guidelines
87
and implicit (at last as far as government is concerned) in the
Procurement and Consultant Guidelines provision that allows for sanctions of firms and
individuals.

November 15, 2010

33

129. Nevertheless, it is our view that government officials should be susceptible to
sanction when they act in their individual rather than their official capacity. For example,
a government official that controls a firm that bids on Bank financed contracts should be
liable to sanction for any Sanctionable Practice that he or she engages in as head of the
bidding firm. In such situations, the notions of comity and respect for sovereignty that
underlie that Banks policy do not apply. There is already precedent for this sort of
distinction in the Banks current practice: state-owned enterprises are liable to sanction
when they operate autonomously and participate in bidding for Bank financed contracts,
while they are not liable to sanction (and nor are they able to bid on Bank financed
contracts) if they are essentially arms of government acting within their own country.
88

130. We would also point out, however, that cases against government officials must
not run afoul of the political prohibition contained in the Article of Agreement.
89
Such
cases, particularly if pursued against high officials or political appointees, have the
potential of entangling the Bank in the political affairs of the member country concerned,
or give the appearance thereof. We would therefore strongly advise INT to seek LEGs
advice and to consult with the relevant Region before launching any such case.
November 15, 2010

34

NOTES

1
Any advice we may have as to the application of the principles articulated in this Advisory Opinion to
specific cases will be communicated separately.
2
IBRD Articles of Agreement, Article III, Section 5(b) (as amended effective February 16, 1989).
3
See Report Concerning the Debarment Process of the World Bank, August 14, 2002 by Dick
Thornburgh et al [hereinafter Thornburgh Report].
4
Reform of the World Banks Sanctions Process, dated June 28, 2004 [Board Paper R2004-0025/2].
5
See, e.g., Sanctions Reform: Proposal for Early Temporary Suspension dated April 15, 2009 (AC2009-
0120); see also Sanctions Reform: Proposal for Early Temporary Suspension dated June 5, 2009
(AC2009-0025/2).
6
See office memorandum dated, April 30, 2008, from Scott White, Acting Vice President and General
Counsel, to Pascale Dubois, Evaluation and Suspension Officer [hereinafter the 2008 LEG memorandum]
paragraph 8.
7
Sanctions Board Statute, Article III.
8
Sanctions Procedures Section 1.01(c) (proposed July 1, 2010 version). The same language is found in
Art. I, Section (c) of the current Sanctions Procedures.
9
Id.
10
A related argument that is sometimes made is that since the legal agreement between the Bank and the
borrower is adopted by the borrower as part of its national law, once a new version of the definitions has
been agreed by a borrower through a legal agreement, it should apply as any new law, to any conduct that
occurs after its adoption. This argument is flawed for a number of reasons, but mainly because (i) not all
borrowers are countries, (ii) not all borrower countries incorporate legal agreements with the Bank into
national law and (iii) in any event, the law in question, by its terms only applies to the implementation of
a particular project.
11
See Thornburgh Report, pp. 10-11, footnote 6 [supra note 3].
12
IFC, MIGA and Bank Guarantees adopted Anti-Corruption Guidelines in 2006 which were, in fact,
interpretative glosses on the definitions of Sanctionable Practices.
13
These were discussed and agreed about a year after the adoption of the Guidelines by a working group
led by LEG and including INT, OPCS and the Asia Region. While still in draft, the Commentaries are a
useful tool for interpretation and, once issued, will be authoritative.
14
See Statute of the International Court of Justice, Article 38, Section 1(c); accord WBAT (de Merode,
Case No. 1). See also ILOAT Judgment No. 2906 where the Tribunal applied general principles of law
where the Service Regulations did not provide any specific provision governing the disputed issue.
15
In some Notices that we have seen, there is no argument to support the assertion that a particular theory
of liability is grounded in general principles of law, it is simply taken as a given. This is, as the saying goes,
putting the proverbial cart before the horse.
16
The benchmark jurisdictions for this exercise could be (as we have done to benchmark various
sanctions reforms) the US, UK, France, Germany and possibly China.
17
See Statute of the International Court of Justice, Article 38, Section 1.
18
2008 LEG memorandum, paragraph 8 [supra note 6].
November 15, 2010

35

19
In this connection, LEG will shortly issue a manual [hereinafter the Sanctions Manual] consolidating
and codifying all guidance materials on the sanctions process.
20
Thornburgh Report, pg. 27, footnote 20 [supra note 3].
21
The Sanctions Board Statute provides that the Chair will instruct the Board on matters not covered by
either the Statute or the Procedures. However, this does not preclude the Chair seeking LEGs advice.
Moreover, it does not apply to situations where the matter is covered, but the meaning and application of
the legal framework is not clear.
22
The 2006 Sanctions Reform Board Paper stated with respect to the effective date that: [t]he foregoing
reforms will apply to all IBRD and IDA financed investment projects, as well as PRG operations, for
which the Project Concept Note is issued on or after October 15, 2006. There was a clear intention that the
reforms contained in the 2006 Sanctions Reform Board Paper would not be retroactive. See Sanctions
Reform: Expansion of Sanctions Regime Beyond Procurement and Sanctioning of Obstructive Practices
(R2006-0149/4) dated August 1, 2006, paragraph 44.
23
We have provided a full treatment of jurisdictional issues in the Sanctions Manual.
24
In cases where, as sometimes still happens, the original legal agreement does not include such provisions,
jurisdiction may still be established by a showing, either through a formal amendment of the legal
agreement or through an exchange of communication or course of dealing, a subsequent agreement
between the Bank and the borrower as to the Banks right to sanction.
25
Thus, in the case of supply contracts with UN agencies financed through IBRD loans or IDA credits,
where the UN agency itself is exempted from Bank sanctions, the Bank asserts the right to investigate and
sanction third parties, notwithstanding the lack of any consent thereto by those third-parties, based on the
fact that the Procurement, Consultant and Anti-Corruption Guidelines nevertheless apply to the project.
26
See Sanctions Reform: Proposals for Improving the Sanctions Process (AC2009-0075) dated July
17, 2009.
27
See, e.g., Section 9.02 (b) of the IDA General Conditions which provides that: The representative so
designated by the Recipient or person so authorized by such representative may agree to any modification
or amplification of the provisions of the Financing Agreement on behalf of the Recipient by written
instrument executed by such representative or authorized person; provided that, in the opinion of such
representative, the modification or amplification is reasonable in the circumstances and will not
substantially increase the obligations of the Recipient under the Financing Agreement. The Association
may accept the execution by such representative or other authorized person of any such instrument as
conclusive evidence that such representative is of such opinion. Similar provisions are set out in the IBRD
General Conditions and the Standard Conditions for Grants.
28
Guidelines Procurement Under IBRD Loans and IDA Credits, Section 1.1.
29
The best evidence rule is a common law rule of evidence which can be traced back at least as far as the
18th century to the effect that no evidence was admissible unless it was "the best that the nature of the case
will allow". The general rule is that secondary evidence, such as a copy or facsimile, will be not admissible
if an original document exists, and is not unavailable due to destruction or other circumstances indicating
unavailability.
30
For example, in one recent case mentioned to us by OES, the issue in question related to a bid of which a
copy was not available. INT presented circumstantial evidence that tended to show that the bid had been
submitted by the Respondent. OES found that there was sufficient evidence to show that the bid had been
submitted, but, without the bid itself (or a copy), the evidence was insufficient to support the allegation that
the bid had been fraudulent in nature.
31
See Thornburgh Report, pg. 49, supra note 3.
November 15, 2010

36

32
This guidance has now been incorporated into Chapter 8 of the Sanctions Manual.
33
However, INT is required under the Section 3.02 of the Sanctions Procedures to present to the EO and
the Sanctions Board all relevant evidence, including exculpatory and mitigating evidence. Therefore, some
balancing of contrary evidence may still be required.
34
The matter pre-dates current INT and SLU management and the establishment of OES and the Sanctions
Board. The issue was handled by Duncan Smith for INT and Hassane Cisse and Frank Fariello for LEG.
35
See Report to the Board from the Audit Committee dated May 24, 2006 (AC2006-0057). Paragraph 8
states that: With respect to the issue of intent, management elaborated saying that a degree of intent can be
inferred from the requirement in the definition that the corrupt act was done for a purpose, i.e., to influence
improperly the actions of another party. That is sufficient for an administrative process, and management
believes that it would be inappropriate for the Bank to raise the bar to a criminal or quasi-criminal standard
in an administrative process. Furthermore, the Procurement Guidelines do not specifically refer to intent
which is implicit in the context.
36
As explained at greater length below, the actual degree of awareness of the risk depends on the area of
law. Criminal recklessness requires actual, subjective knowledge of the risk, while in tortious recklessness
and civil fraud, recklessness may be established through inference, willful blindness or simple
cluelessness as to the risk.
37
The American Law Institutes Model Penal Code also recognizes a category of mens rea of knowingly
where the offender does not necessarily desire the result, but is practically certain that it will ensue from his
actions.
38
Strict liability is used, for example, on social utility grounds in cases of statutory rape or drunk driving
or, on efficiency grounds, for minor offenses (e.g., parking violations). See, e.g., the Sexual Offenses Act of
2003.
39
For example, the offender places an explosive on a bus in order to kill the driver. The murder of the
driver will be considered dolus directus, while the damages of the bus are considered as dolus in the second
degree.
40
The civil law also recognizes a form of negligence known in Spanish as culpa con representacin: The
offender accepts that as a result of his behavior an offence may occur. The offender is aware of the possible
result of his actions and goes ahead anyway. However, the offender has the confidence that because of his
personal skills the result will not be achieved. Here again, in most cases, this would probably be considered
as a form of recklessness under common law. The limit between the dolus eventualis and the culpa con
representacin is very narrow and in most cases theoretical.
41
E.g., The German Strict Liability Act, under which strict liability is imposed for death, personal injury
and property damage caused through the operation of a railway. Also, Pharmaceutical Products Act
renders producers of medical products liable if death or serious personal injury is caused by the use of the
product as prescribed, provided the product's harmful effects go beyond those which are to be tolerated
according to the current state of medical science. There of course is the Product Liability Act, too where
the producer's liability for personal injury and property damage caused by a defective product regardless of
negligence is established.
42
It bears mentioning that the requirement of direct involvement, which was contained in the original
guidance note that accompanied the Sanctioning Guidelines, has now been replaced with guidance that
focuses on the between culpability and responsibility, and we will base our analysis on that distinction.
43
See 2008 LEG memorandum [supra note 6].
44
Id.
November 15, 2010

37

45
See Guidelines: Procurement under IBRD Loans and IDA Credit and Guidelines: Selection and
Employment of Consultants by World Bank Borrowers: Proposed Modification, dated June 25, 2003
(R2003-0129; IDA/R2003-0152) discussed at the Board meeting of July 17, 2003.
46
Dick Thornburgh, Ronald L. Gainer, Cuyler H. Walker, Report to Shengman Zhang, Managing Director
and Chairman of the Oversight Committee on Fraud and Corruption, Concerning Mechanisms to Address
Problems of Fraud and Corruption (January 21, 2000 (Rev.)), page 57.
47
Thornburgh Report, page 31 [supra note 3].
48
Furthermore, in a 2006 Audit Committee paper on sanctions reform, it was observed that the words
directly or indirectly in the current definition of corrupt practice could be cited to capture participatory
acts such as aiding and abetting, while conspiracy and complicity were captured in the definition of
collusive practice.

See Sanctions Reform: Background Note on the Definitions of Fraud and Corruption
(AC2006-0040) discussed at the May 24, 2006 meeting of the Audit Committee.
49
Modem criminal law has come to apply the notion of indirect perpetration even where the direct and
physical perpetrator is criminally responsible ("perpetrator behind the perpetrator").This is especially
relevant if crimes are committed through an organized structure of power in which the direct and physical
perpetrator is nothing but a cog in the wheel that can be replaced Immediately. Since the identity of the
direct and physical perpetrator is irrelevant, the control and, consequently, the main responsibility for the
crime committed shifts to the persons occupying a leading position in such an organized structure of power.
See, e.g., German Strafgesetbuch, Sec. 25 (1): Whoever commits the crime himself or through another
person shall be punished as perpetrator.
50
Co-perpetration generally requires joint functional control over a crime. Co-perpetrators must pursue a
common goal, either through an explicit agreement or silent consent, which they can only achieve by
coordinated action and shared control over the criminal conduct. See, e.g., the German Strafgesetbuch,
Section 25 (2): If a number of persons commit the crime jointly, each shall be punished as perpetrator (co-
perpetrator).
51
Such treatment is provided for in the Procurement Guidelines. See Guidelines, Procurement under IBRD
Loans and IDA Credits, Appendix III, paragraph 6 (Revised, October 2006).
52
Compare, however, US criminal law, which requires for an employer to be criminally liable for the acts
of its employee, it must be shown that the employees acts were within the scope of his or her employment
and undertaken in the course of the employers business, or that the acts were habitually done in the course
of the business. Typically, this means showing that the employee was authorized by management; passive
acquiescence or even knowledge of the acts is insufficient standing alone to establish liability
53
Prosecutor v Dusko Tadic, Case No.: IT-94-1-A, Judgment, 15 July 1999.
54
Crim. 24 aout 1827, B. no. 224 (France), Pinkerton v. US (328 U.S. 640) (US).
55
See Pinkerton v. US (328 U.S. 640).
56
See International Tribunal for Rwanda (ICTR). Nkatirutimana Appeal Judgment, paras. 463, 468.
57
The third, systemic form of joint criminal enterprise is a variant of the basic form, characterized by the
existence of an organized system of ill-treatment. An example is extermination or concentration camps, in
which the prisoners are killed or mistreated pursuant to the joint criminal enterprise. This form of joint
criminal enterprise, however, appears to have little or no application to the Banks sanctions regime.
58
See In re Stashynskij (BGHSt 18, 87) (German).
59
It may actually be more precise to call this a scienter requirement because, as explained below, it
relates essentially the state of knowledge of the accused party.
November 15, 2010

38

60
See, e.g., paragraph 1.14 of the Procurement Guidelines. Of course, these two areas of concern are not
mutually exclusive, but the point is that a failure of professional conduct that does not rise to the level of an
ethical failure was not intended to be sanctionable under our current sanctions regime. There are those who
contend that this should not be the case, and thatas in the case of many national debarment regimes
breach of contract and poor performance, for example, should be grounds for debarment. Whatever the
merits of that position, which we should consider in the course of the upcoming review of the sanctions
regime, it is clear that our current system was not intended to extend that far.
61
This theory was invoked in the most recent cohort of Sanctions Board decisions. See, e.g., Case No. 105
(India). decided on September 20, 2010 (SB Case No. 36).
62
See Joshua Dresler, "Understanding Criminal Law " Fifth edition, pg. 135 (March 2006).
63
See State v. Engle, 731 N.W. 2d 852 (Minn. Ct. App. 2007).
64
See Dorsey v. U.S., 902 A. 2d 107 (D.C. 2006), but see also People v. Deskins, 927 P. 2d 368 (Colo.
1996), discussed in detail below.
65
Negligence is generally a tort law concept, but certain crimes require only criminal negligence:
inadvertent risk-taking that represents a gross deviation from a reasonable standard of care. The offender
takes a substantial and unjustifiable risk, but, unlike a case of recklessness, he/she may not be aware of the
risk. Criminal negligence is controversial for allowing the punishment of persons who have acted without
malice, but is justified on utilitarian ground of general deterrenceand sometimes a sense of natural
justice: the level of negligence should be so great that it would be shocking to allow the actors lack of
awareness to excuse his actions under the circumstances.
66
Securities Exchange Act 1034- Rule 10 b-5- Employment of Manipulative and Deceptive Devices.
67
See Ernst & Ernst v. Hochfelder, 425 U.S. 185, 193-194 (1976).
68
See 15. U.S.C. S 78m (b) (2).
69
See US v. Jensen, 532 F. Supp. 2d 1187 (2007).
70
31 U.S.C 3729 (b)(1)(A)(ii),(iii).
71
See Miller v. U.S., 213 Ct. Cl. 59, 550 F.2d 17, 23 (1977).
72
See U.S. v. Krizek et al., 111 F.3d 934 (D.C. Cir. May 2, 1997).
73
See Senate Judiciary Committee, False Claims Amendments Act of 1986, S. Rep. No. 99-345, at 21
(1986); see also Crane Helicopter Servs., Inc. v. United States, 45 Fed. Cl. 410, 433 (1999) ([The FCA's
knowing standard was designed to address] the ostrich-like refusal to learn of information which an
individual, in the exercise of prudent judgment, had reason to know' and to reach those who ignore
obvious warning signs.)
74
See Miller, 550 F.2d 17, 23.
75
Similar allowance of inference can be found in various conventions on money laundering, e.g., the
Vienna Convention, the Palermo Convention, the FATF Recommendations.
76
To establish simple negligence, it is often said that the risk of harm must have been foreseeable to the
reasonable person. Without foreseeability, not even simple negligence can be established. Foreseeability
may be judged objectively, through the existence (or lack thereof) of red flags that would put a reasonable
person on notice of the risk of harm. To establish recklessness, of course, the likelihood must be much
higher than simple foreseeability.
77
Blacks Law Dictionary (8
th
Ed. 2004).
78
This is not to say that the conduct should not be considered contextually. Without trying to draw too fine
a line, the standard does ask what the proverbial reasonable person would do under the circumstances, so
November 15, 2010

39

the context in which the act takes place is relevant. But this is not quite the same as adopting a variable
reasonable person standard, since different people react differently under the same circumstances.
79
Under the Banks standard bidding documents, bid signatories do attest to the accuracy of bid
information to the best of [their] knowledge, but there is no requirement as to the due diligence (if any)
that should underlie the attestation.
80
W. Prosser, Handbook of the Law of Torts 34, at 183-84 (4th ed. 1971) (footnotes omitted). The Model
Penal Code distinguishes between acting recklessly and acting negligently according to whether a person
"consciously disregarded" or simply "should be aware of" a substantial and unjustifiable risk. Model Penal
Code 2.02, reprinted in 10 Uniform Laws Ann. at 465.
81
57A AmJur2d 281.
82
See Section 9.04 of the Sanctions Procedures and guidance in Annex B to the Audit Committee paper on
Sanctions Reform: Phase III (AC2010-0047) dated May 12, 2010.
83
See Section IV of the Sanctioning Guidelines; see also Case No. 102 (India) decided September 20, 2010
(Sanctions Board Case No. 38).
84
See 37 Am Jur 2d Fraud and Deceit 26.
85
See Sanctions Reform: Background Note on Definitions of Fraud and Corruption (AC2006-0040) dated
May 23, 2006 and the related slideshow (AC2006-0040/1) dated June 7, 2006.
86
See India case [supra note 83], where a Respondent signed an affidavit attesting to the accuracy of the
information provided in the bid and accepting personal responsibilityincluding possible sanctionsin the
event that the information were to prove inaccurate.
87
See IBRD/IDA Anti-Corruption Guidelines, paragraph 11(a) and the related footnote 14.
88
See Procurement Guidelines, paragraph 1.08(c), Consultant Guidelines, paragraph 1.11(b), and
IBRD/IDA Anti-Corruption Guidelines, paragraph 11(a), footnote 14.
89
The Articles of Agreement of the IBRD contain a provision that states that [t]he Bank and its officers
shall not interfere in the political affairs of any member; nor shall they be influenced in their decisions by
the political character of the member or members concerned. (Article IV, Section 10); see also IDA
Articles of Agreement, which contain an identical provision (Article V, Section 6).

Advisory Opinion

Cargado por

Información del documento

Derechos de autor

Formatos disponibles

Compartir este documento

Compartir o incrustar documentos

Opciones para compartir

¿Le pareció útil este documento?

¿Este contenido es inapropiado?

Copyright:

Formatos disponibles

Advisory Opinion

Cargado por

Copyright:

Formatos disponibles

Legal Vice Presidency

The World Bank

También podría gustarte